Assessment of Post-Exertional Malaise (PEM) in Patients with Myalgic Encephalomyelitis (ME) and Chronic Fatigue Syndrome (CFS). Holtzman et al. (2019)

New study on PEM by the research team of Leonard Jason. I'm sure many on this forum will have filled in the online questionnaire:

 

They offered an extensive, online questionnaire on PEM to ca. 1500 patients. Obviously, the ME/CFS diagnosis was not confirmed by a clinician of the research team, patients simpled reported to have been diagnosed with ME or CFS. The paper is open-access: https://www.mdpi.com/2075-4418/9/1/26

1) One interesting finding was that more than 80% of patients endorsed several non-exertion triggers of PEM such as emotional events, noise or sensory overload. Other items such as heat or watching a video also scored high: it seems like anything that takes away energy, can trigger PEM.

2) More than 90% of patients experienced a delayed onset of PEM. So that’s a thing. It also seems that PEM can last a long time. 30% said that the recovery period can last 1-6 months. Two-thirds said they suffered a crash that never resolved.

3) Most patients said that pacing helped but that it was only moderately or mildly effective. In one of the most intriguing findings, only 6% said that pacing allows them to completely avoid symptom exacerbation.

4) Exercising within limits didn’t seem to work for most patients. Only 37% of subjects reported being able to exercise a little without PEM-related symptoms, as long as they stay within certain limits. This corresponds with earlier findings by the research team of Nijs: exercise limits cannot completely prevent PEM: https://www.ncbi.nlm.nih.gov/pubmed/18441039

 

Looks like this could be another useful one for the NICE guideline review @adambeyoncelowe @Keela Too @saranbonser

An example,

 

And I would count myself in that 37%... 3 years ago. Not always, but I was able to until then. Now I have bottomed out, have not had a significant remission since then. So that is likely a high estimate as many who can for some time will end up not being able to. Whether this is a natural course or aggravated by pushing through (which I did) is impossible to tell, but this is likely a high estimate.

 

I'm delighted to see this study.. A true reflection of this awful illness.

 

Is it really likely to count as evidence, given there was no guarantee (as I understand) of a confirmed diagnosis?

 

Well, it should carry as much weight as any other patient survey.

 

I can't think of many studies that have been totally physician diagnosed, but perhaps I have read too many paediatric ones where it is really fatigue focused.
Even physician diagnosed may quote Oxford so are not limited to an ME cohort.
This is where diagnostic criteria matter.

 

Anything that depends on diagnostic coding to identify ... there are a few studies like this out there. Ideally, you'd get an ME expert or two to confirm that the person had been diagnosed correctly, or you'd ask them to diagnose with your chosen research criteria. Clinical criteria are looser in general anyway, because the goal is to be able to do something that helps the patient.

Sometimes that's getting them on disability or getting them a med they need, and that requires certain diagnoses, which can mean the chart is, if not wrong, per se, then selective... I have had doctors ask me if I "wanted" specific diagnoses on my chart or not!

 

So they've done two things:
1/ Develop a 'comprehensive PEM instrument' with the help of significant patient input.
2/ Get a lot of self-reported data using said 'comprehensive PEM instrument'.

Yes, I participated in both phases, so it's great to see the outcome of it all published.

I think using patient input they way they did in phase 1 - during the development of study questions and instruments - is fantastic and should be done much more often. Of course that requires some extra effort initially by the researchers. I'm sure I drove them mad with my constant feedback about this question being ambiguous and that one being liable to be misunderstood. I wasn't 100% happy with the final questionnaire but it was a lot better than anything else out there I've seen so far. And to be fair it is excruciatingly difficult to formulate unambiguous questions that lead to answers that actually measure what you think you're measuring.

The second phase collected some very interesting data that, I'm sure, paints a much more recognisable picture of PEM than we usually get.

I, too, am worried that it won't be taken as seriously as it could because of the self-reported diagnosis issue, self-reported everything really (not that this is any different from much other ME/CFS research but if somebody wants to attack the study a minor issue like the pot calling the kettle black won't stop them).

It would have been better had they done the second phase with a separate cohort with confirmed diagnoses, plus one or more control/comparison groups for good measure. But that would have cost a lot more (there was no external funding) and maybe some other group will now use the questionnaire on other cohorts.

 

We should probably see this as the first, preliminary test of the questionnaire, which gives us some interesting data and ideas but was never meant to be used as facts. As far as I could tell the authors make no claims that the data means anything other than that the questionnaire seems to be getting at something previous versions have not. Apologies if I missed something!

Hopefully they will try to confirm the results with a different cohort later.

 

I’m sure this could be used to compare the response patterns of various different criteria used during diagnosis. (Eg SEID, CCC, ICC etc etc.). Comparing such responses could be enlightening.

It might be also interesting to compare the responses of patients diagnosed by different centres. I have a feeling that criteria (whichever set are used) might get applied in different ways by different places.

Would this pick up that type of cohort difference?
Could a smaller set of questions be found from all this, that would discern important groupings with less effort from the patients?

All interesting to ponder?

 

We live in an age of information and data mining. Asking lots of people who think they have CFS or ME or ME/CFS what symptoms they get might be able to find patterns which distinguish different illnesses or at least presentations of one illness.

Our biggest problem is that medical authority does not know what our disease involves. They think it is fatigue (because it suits them) but it is not.

Fatigue as a symptom is common in many diseases but the CAUSE varies immensely. In heart disease the heart cannot keep enough blood going, in COPD there is not enough oxygen being breathed in, in MS and RA fatigue is often because of the sheer difficulty of moving. But for us, we appear fatigued because we have a broken cellular respiration system.

I simplify, but fatigue is best treated by looking at the cause of that fatigue but the medical establishment don't consider the cause of ours and it is time they did.

 

Not much different from the trials used as evidence for the psychosocial model. Most of them have an unknown number of ME patients, if any. There is no reason why that standard should be selective. It means both sets of studies have limited reliability. Either they exclude all studies with this flaw or they accept them all.

PACE included participants with major depression and somatization diagnoses that would be exclusionary criteria for an ME diagnosis (at least in research). Their studies are all a grab-bag of near random participants who may or may not have ME. They have no leg to stand on as far as eligibility of participants go. There is no reason to accept a standard in which they can do whatever the F they want on their trials but demand requirements they don't meet themselves of studies that don't follow their ideological framework.

 

Oxford criteria are also self-reported. There is no reason why that point should be ignored. Technically this selection is actually on firmer ground as ME criteria are much more stringent than literally-the-vaguest-possible Oxford criteria. The presence of a physician in the process is irrelevant beyond writing down that self-diagnosis and having no differential diagnosis to offer.

A physician cannot confirm or falsify the presence of fatigue, as it is a subjective symptom. By definition, and in practice, Oxford criteria are self-reported. As are any questionnaires used, whether they are based on a competent definition of ME or the generic concepts of fatigue. In practical terms, Jason's participants are much more likely to be ME patients than any of the psychosocial studies, which we know include non-ME participants by design (major depressive, other psychiatric diagnoses, somatization, mood disorders, etc.).

 

Yes, this is the most useful role of the survey. The causative factor(s) aren't precisely known. So, at least the wide-ranging definitions of PEM are an opening to reasses and redefine PEM with an eye to discerning differences clinically which may point to different causative factors or innate individual reactions.

 

The ones we expose as bad science you mean ...

 

My sentiments entirely Ravn. And I had plenty to say to them about those things too. So good that they actually bothered to ask & then listen.

I hope it'll be useful for this too, because..

Mithriel you are sooo right, this is the part i find so mindbendigly frustrating that even those in authority that have a biomedical approach, still dont seem to fully grasp the true clinical picture, let alone the others.

yes... you make an uncomfortable, but very good point

 

Questionnaires are definitely preliminary research. Preliminary research is a necessary step, but no serious research with actionable conclusions can avoid having objective measures in medicine.

This study is tier 1 research. So is PACE, despite its tier 3 price tag and outlandish promises. They are on equal footing in that regard. Either they accept both or they accept neither. There is not a single feature of PACE that makes it more authoritative other than eminence. Any justification to exclude it would exclude PACE on the same grounds. That's not a bug, that's a feature.